barto_exp1_10partition_modelo_msl9000

This model is a fine-tuned version of vgaraujov/bart-base-spanish on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4960
• Model Preparation Time: 0.0056
• Bleu Msl: 0
• Bleu 1 Msl: 0.8086
• Bleu 2 Msl: 0.7441
• Bleu 3 Msl: 0.6848
• Bleu 4 Msl: 0.5835
• Ter Msl: 24.1228
• Bleu Asl: 0
• Bleu 1 Asl: 0.9818
• Bleu 2 Asl: 0.9669
• Bleu 3 Asl: 0.9485
• Bleu 4 Asl: 0.9264
• Ter Asl: 2.0278

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0001
• train_batch_size: 32
• eval_batch_size: 64
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 30
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Model Preparation Time	Bleu Msl	Bleu 1 Msl	Bleu 2 Msl	Bleu 3 Msl	Bleu 4 Msl	Ter Msl	Bleu Asl	Bleu 1 Asl	Bleu 2 Asl	Bleu 3 Asl	Bleu 4 Asl	Ter Asl
No log	1.0	225	0.3301	0.0056	0	0.8146	0.7241	0.6553	0.5580	22.9407	0	0.9593	0.9326	0.9053	0.8755	5.0133
No log	2.0	450	0.2479	0.0056	0	0.8706	0.8179	0.7650	0.6741	15.0576	0	0.9609	0.9353	0.9085	0.8778	4.8614
0.4382	3.0	675	0.2646	0.0056	0	0.8355	0.7407	0.6656	0.5501	18.8663	0	0.9732	0.9560	0.9365	0.9126	3.1903
0.4382	4.0	900	0.2865	0.0056	0	0.8395	0.7867	0.7364	0.6492	16.0319	0	0.9729	0.9547	0.9344	0.9102	3.4182
0.0856	5.0	1125	0.3001	0.0056	0	0.6942	0.6154	0.5419	0.4526	25.6864	0	0.9761	0.9604	0.9419	0.9201	2.8864
0.0856	6.0	1350	0.2791	0.0056	0	0.8284	0.7660	0.7068	0.6120	19.0434	0	0.9738	0.9565	0.9370	0.9132	3.3422
0.0492	7.0	1575	0.3060	0.0056	0	0.8044	0.7356	0.6741	0.5849	19.8406	0	0.9735	0.9547	0.9339	0.9082	3.2662
0.0492	8.0	1800	0.3230	0.0056	0	0.7931	0.6757	0.5879	0.4854	25.5979	0	0.9745	0.9574	0.9387	0.9161	3.1523
0.0344	9.0	2025	0.2935	0.0056	0	0.8312	0.7728	0.7138	0.6320	18.9548	0	0.9716	0.9539	0.9343	0.9105	3.4182
0.0344	10.0	2250	0.2862	0.0056	0	0.7703	0.6791	0.6071	0.5190	23.3835	0	0.9722	0.9556	0.9369	0.9141	3.2662
0.0344	11.0	2475	0.2891	0.0056	0	0.8084	0.7362	0.6689	0.5775	20.5492	0	0.9764	0.9603	0.9414	0.9183	2.7345
0.0233	12.0	2700	0.3214	0.0056	0	0.7989	0.7230	0.6543	0.5484	23.7378	0	0.9787	0.9637	0.9467	0.9257	2.5066
0.0233	13.0	2925	0.3377	0.0056	0	0.8345	0.7662	0.7035	0.6164	19.6634	0	0.9764	0.9613	0.9441	0.9226	2.8105
0.0171	14.0	3150	0.3142	0.0056	0	0.8276	0.7770	0.7231	0.6420	16.9176	0	0.9783	0.9635	0.9463	0.9251	2.5826
0.0171	15.0	3375	0.3416	0.0056	0	0.7555	0.6759	0.6036	0.5096	22.6749	0	0.9793	0.9651	0.9487	0.9280	2.5446
0.012	16.0	3600	0.3114	0.0056	0	0.8309	0.7624	0.6982	0.6022	18.3348	0	0.9745	0.9592	0.9417	0.9202	2.8864
0.012	17.0	3825	0.3250	0.0056	0	0.8369	0.7608	0.6953	0.6006	18.9548	0	0.9761	0.9595	0.9412	0.9183	3.1523
0.0097	18.0	4050	0.3011	0.0056	0	0.8052	0.7281	0.6589	0.5632	20.9920	0	0.9761	0.9593	0.9405	0.9174	3.0763
0.0097	19.0	4275	0.3326	0.0056	0	0.8163	0.7387	0.6672	0.5704	19.7520	0	0.9764	0.9605	0.9425	0.9202	2.8485
0.0065	20.0	4500	0.3216	0.0056	0	0.7911	0.7136	0.6439	0.5477	22.4092	0	0.9787	0.9630	0.9452	0.9230	2.6586
0.0065	21.0	4725	0.3671	0.0056	0	0.7975	0.7124	0.6407	0.5420	21.9663	0	0.9761	0.9603	0.9424	0.9200	2.8864
0.0065	22.0	4950	0.3547	0.0056	0	0.8034	0.7246	0.6558	0.5551	21.4349	0	0.9761	0.9596	0.9414	0.9189	2.9624
0.0051	23.0	5175	0.3530	0.0056	0	0.8076	0.7277	0.6583	0.5611	20.9035	0	0.9758	0.9592	0.9407	0.9179	3.0004
0.0051	24.0	5400	0.3531	0.0056	0	0.8041	0.7369	0.6732	0.5764	20.7263	0	0.9761	0.9598	0.9415	0.9189	2.8105
0.0035	25.0	5625	0.3439	0.0056	0	0.7976	0.7296	0.6670	0.5738	21.1692	0	0.9758	0.9596	0.9418	0.9196	2.8485
0.0035	26.0	5850	0.3522	0.0056	0	0.8124	0.7455	0.6837	0.5891	19.3091	0	0.9764	0.9605	0.9425	0.9204	2.8105
0.0034	27.0	6075	0.3768	0.0056	0	0.8062	0.7375	0.6735	0.5790	20.1063	0	0.9764	0.9605	0.9424	0.9201	2.8105
0.0034	28.0	6300	0.3592	0.0056	0	0.8070	0.7379	0.6738	0.5792	20.0177	0	0.9764	0.9602	0.9417	0.9189	2.8864
0.0026	29.0	6525	0.3571	0.0056	0	0.8114	0.7430	0.6802	0.5855	19.4863	0	0.9767	0.9606	0.9421	0.9194	2.8485
0.0026	30.0	6750	0.3563	0.0056	0	0.8106	0.7421	0.6790	0.5847	19.5748	0	0.9764	0.9602	0.9417	0.9189	2.8864

Framework versions

• Transformers 4.51.1
• Pytorch 2.6.0+cu124
• Datasets 3.5.0
• Tokenizers 0.21.1